Development of face recognition gates program

2020-12-22 14:18:03 fandoukeji

Development of face recognition gates program

Application background

The average daily passenger flow of the railway station is very large, especially during holidays, the passenger flow doubles. Ticket inspectors have to face a huge flow of people for identity verification every day, requiring a large investment of manpower and material resources, manual inspection takes a long time, and manual inspection will inevitably lead to fatigue, resulting in misjudgment or missed judgment, which leads to those who do not hold their own documents. People can also pass, causing adverse effects.

Using the current advanced face recognition technology in the station face recognition gates, passengers can enter the station by "swiping their faces". The process only takes 3 to 5 seconds, which shortens the entry time, reduces the pressure on ticket inspectors, and avoids artificial eyes. Determine the various drawbacks, and quickly determine whether it is the person and the fugitive suspect by comparing ID photos and faces in real time, which improves the security level, and can also effectively control the travel records of passengers and improve the train station The safety management mechanism creates a safe environment in the railway station and provides passengers with humane and high-quality services.


A camera is installed in the face recognition gate of the “face-swiping” entry channel. When the passenger approaches the machine, the camera will lock the passenger’s facial information and compare it with the photo in the ID card chip. The ticket information matches the face. If it matches the ID photo, the gate will automatically release it.

Program features

RK3288 uses a super Cortex-A17 quad-core processor with a main frequency of 1.8GHz, 2G memory, 8G eMMc, and powerful hardware resources to provide a strong guarantee for the efficient operation and calculation of the face recognition algorithm;

Advanced Internet of Things architecture technology, each device has an independent IP, realizes pure network independent operation across network segments, and the network is simple and fast;

Automatic data upload, supporting real-time upload of verifier’s identity information and verification results to the back-end management system;

Rich development interfaces can help face recognition algorithm software vendors to quickly realize the intelligentization of face recognition solutions.